DeepFrag-k: a fragment-based deep learning approach for protein fold recognition

BACKGROUND

One of the most essential problems in structural bioinformatics is protein fold recognition. In this paper, we design a novel deep learning architecture, so-called DeepFrag-k, which identifies fold discriminative features at fragment level to improve the accuracy of protein fold recognition. DeepFrag-k is composed of two stages: the first stage employs a multi-modal Deep Belief Network (DBN) to predict the potential structural fragments given a sequence, represented as a fragment vector, and then the second stage uses a deep convolutional neural network (CNN) to classify the fragment vector into the corresponding fold.

RESULTS

Our results show that DeepFrag-k yields 92.98% accuracy in predicting the top-100 most popular fragments, which can be used to generate discriminative fragment feature vectors to improve protein fold recognition.

CONCLUSIONS

There is a set of fragments that can serve as structural "keywords" distinguishing between major protein folds. The deep learning architecture in DeepFrag-k is able to accurately identify these fragments as structure features to improve protein fold recognition.

Relevance of Machine Learning Techniques and Various Protein Features in Protein Fold Classification: A Review

Background:

Protein fold prediction is a fundamental step in Structural Bioinformatics. The tertiary structure of a protein determines its function and to predict its tertiary structure, fold prediction serves an important role. Protein fold is simply the arrangement of the secondary structure elements relative to each other in space. A number of studies have been carried out till date by different research groups working worldwide in this field by using the combination of different benchmark datasets, different types of descriptors, features and classification techniques.

Objective:

In this study, we have tried to put all these contributions together, analyze their study and to compare different techniques used by them.

Methods:

Different features are derived from protein sequence, its secondary structure, different physicochemical properties of amino acids, domain composition, Position Specific Scoring Matrix, profile and threading techniques.

Conclusion:

Combination of these different features can improve classification accuracy to a large extent. With the help of this survey, one can know the most suitable feature/attribute set and classification technique for this multi-class protein fold classification problem.

Classification of Protein Structure Classes on Flexible Neutral Tree

Accurate classification on protein structural is playing an important role in Bioinformatics. An increase in evidence demonstrates that a variety of classification methods have been employed in such a field. In this research, the features of amino acids composition, secondary structure's feature, and correlation coefficient of amino acid dimers and amino acid triplets have been used. Flexible neutral tree (FNT), a particular tree structure neutral network, has been employed as the classification model in the protein structures' classification framework. Considering different feature groups owing diverse roles in the model, impact factors of different groups have been put forward in this research. In order to evaluate different impact factors, Impact Factors Scaling (IFS) algorithm, which aim at reducing redundant information of the selected features in some degree, have been put forward. To examine the performance of such framework, the 640, 1189, and ASTRAL datasets are employed as the low-homology protein structure benchmark datasets. Experimental results demonstrate that the performance of the proposed method is better than the other methods in the low-homology protein tertiary structures.

Recent Progress in Machine Learning-Based Methods for Protein Fold Recognition

Knowledge on protein folding has a profound impact on understanding the heterogeneity and molecular function of proteins, further facilitating drug design. Predicting the 3D structure (fold) of a protein is a key problem in molecular biology. Determination of the fold of a protein mainly relies on molecular experimental methods. With the development of next-generation sequencing techniques, the discovery of new protein sequences has been rapidly increasing. With such a great number of proteins, the use of experimental techniques to determine protein folding is extremely difficult because these techniques are time consuming and expensive. Thus, developing computational prediction methods that can automatically, rapidly, and accurately classify unknown protein sequences into specific fold categories is urgently needed. Computational recognition of protein folds has been a recent research hotspot in bioinformatics and computational biology. Many computational efforts have been made, generating a variety of computational prediction methods. In this review, we conduct a comprehensive survey of recent computational methods, especially machine learning-based methods, for protein fold recognition. This review is anticipated to assist researchers in their pursuit to systematically understand the computational recognition of protein folds.

ProFold: Protein Fold Classification with Additional Structural Features and a Novel Ensemble Classifier

Protein fold classification plays an important role in both protein functional analysis and drug design. The number of proteins in PDB is very large, but only a very small part is categorized and stored in the SCOPe database. Therefore, it is necessary to develop an efficient method for protein fold classification. In recent years, a variety of classification methods have been used in many protein fold classification studies. In this study, we propose a novel classification method called proFold. We import protein tertiary structure in the period of feature extraction and employ a novel ensemble strategy in the period of classifier training. Compared with existing similar ensemble classifiers using the same widely used dataset (DD-dataset), proFold achieves 76.2% overall accuracy. Another two commonly used datasets, EDD-dataset and TG-dataset, are also tested, of which the accuracies are 93.2% and 94.3%, higher than the existing methods. ProFold is available to the public as a web-server.

The recognition of multi-class protein folds by adding average chemical shifts of secondary structure elements

The recognition of protein folds is an important step in the prediction of protein structure and function. Recently, an increasing number of researchers have sought to improve the methods for protein fold recognition. Following the construction of a dataset consisting of 27 protein fold classes by Ding and Dubchak in 2001, prediction algorithms, parameters and the construction of new datasets have improved for the prediction of protein folds. In this study, we reorganized a dataset consisting of 76-fold classes constructed by Liu et al. and used the values of the increment of diversity, average chemical shifts of secondary structure elements and secondary structure motifs as feature parameters in the recognition of multi-class protein folds. With the combined feature vector as the input parameter for the Random Forests algorithm and ensemble classification strategy, we propose a novel method to identify the 76 protein fold classes. The overall accuracy of the test dataset using an independent test was 66.69%; when the training and test sets were combined, with 5-fold cross-validation, the overall accuracy was 73.43%. This method was further used to predict the test dataset and the corresponding structural classification of the first 27-protein fold class dataset, resulting in overall accuracies of 79.66% and 93.40%, respectively. Moreover, when the training set and test sets were combined, the accuracy using 5-fold cross-validation was 81.21%. Additionally, this approach resulted in improved prediction results using the 27-protein fold class dataset constructed by Ding and Dubchak.

Recognition of 27-class protein folds by adding the interaction of segments and motif information

The recognition of protein folds is an important step for the prediction of protein structure and function. After the recognition of 27-class protein folds in 2001 by Ding and Dubchak, prediction algorithms, prediction parameters, and new datasets for the prediction of protein folds have been improved. However, the influences of interactions from predicted secondary structure segments and motif information on protein folding have not been considered. Therefore, the recognition of 27-class protein folds with the interaction of segments and motif information is very important. Based on the 27-class folds dataset built by Liu et al., amino acid composition, the interactions of secondary structure segments, motif frequency, and predicted secondary structure information were extracted. Using the Random Forest algorithm and the ensemble classification strategy, 27-class protein folds and corresponding structural classification were identified by independent test. The overall accuracy of the testing set and structural classification measured up to 78.38% and 92.55%, respectively. When the training set and testing set were combined, the overall accuracy by 5-fold cross validation was 81.16%. In order to compare with the results of previous researchers, the method above was tested on Ding and Dubchak's dataset which has been widely used by many previous researchers, and an improved overall accuracy 70.24% was obtained.

A Two-Layer Learning Architecture for Multi-Class Protein Folds Classification

Classification of protein folds plays a very important role in the protein structure discovery process, especially when traditional sequence alignment methods fail to yield convincing structural homologies. In this chapter, we have developed a two-layer learning architecture, named TLLA, for multi-class protein folds classification. In the first layer, OET-KNN (Optimized Evidence-Theoretic K Nearest Neighbors) is used as the component classifier to find the most probable K-folds of the query protein. In the second layer, we use support vector machine (SVM) to build the multi-class classifier just on the K-folds, generated in the first layer, rather than on all the 27 folds. For multi-feature combination, ensemble strategy based on voting is selected to give the final classification result. The standard percentage accuracy of our method at ~63% is achieved on the independent testing dataset, where most of the proteins have <25% sequence identity with those in the training dataset. The experimental evaluation based on a widely used benchmark dataset has shown that our approach outperforms the competing methods, implying our approach might become a useful vehicle in the literature.

BS-KNN: An Effective Algorithm for Predicting Protein Subchloroplast Localization

Chloroplasts are organelles found in cells of green plants and eukaryotic algae that conduct photosynthesis. Knowing a protein’s subchloroplast location provides in-depth insights about the protein’s function and the microenvironment where it interacts with other molecules. In this paper, we present BS-KNN, a bit-score weighted K-nearest neighbor method for predicting proteins’ subchloroplast locations. The method makes predictions based on the bit-score weighted Euclidean distance calculated from the composition of selected pseudo-amino acids. Our method achieved 76.4% overall accuracy in assigning proteins to 4 subchloroplast locations in cross-validation. When tested on an independent set that was not seen by the method during the training and feature selection, the method achieved a consistent overall accuracy of 76.0%. The method was also applied to predict subchloroplast locations of proteins in the chloroplast proteome and validated against proteins in Arabidopsis thaliana. The software and datasets of the proposed method are available at https://edisk.fandm.edu/jing.hu/bsknn/bsknn.html.

CE-PLoc: an ensemble classifier for predicting protein subcellular locations by fusing different modes of pseudo amino acid composition

Precise information about protein locations in a cell facilitates in the understanding of the function of a protein and its interaction in the cellular environment. This information further helps in the study of the specific metabolic pathways and other biological processes. We propose an ensemble approach called "CE-PLoc" for predicting subcellular locations based on fusion of individual classifiers. The proposed approach utilizes features obtained from both dipeptide composition (DC) and amphiphilic pseudo amino acid composition (PseAAC) based feature extraction strategies. Different feature spaces are obtained by varying the dimensionality using PseAAC for a selected base learner. The performance of the individual learning mechanisms such as support vector machine, nearest neighbor, probabilistic neural network, covariant discriminant, which are trained using PseAAC based features is first analyzed. Classifiers are developed using same learning mechanism but trained on PseAAC based feature spaces of varying dimensions. These classifiers are combined through voting strategy and an improvement in prediction performance is achieved. Prediction performance is further enhanced by developing CE-PLoc through the combination of different learning mechanisms trained on both DC based feature space and PseAAC based feature spaces of varying dimensions. The predictive performance of proposed CE-PLoc is evaluated for two benchmark datasets of protein subcellular locations using accuracy, MCC, and Q-statistics. Using the jackknife test, prediction accuracies of 81.47 and 83.99% are obtained for 12 and 14 subcellular locations datasets, respectively. In case of independent dataset test, prediction accuracies are 87.04 and 87.33% for 12 and 14 class datasets, respectively.

Protein fold classification with genetic algorithms and feature selection

Protein fold classification is a key step to predicting protein tertiary structures. This paper proposes a novel approach based on genetic algorithms and feature selection to classifying protein folds. Our dataset is divided into a training dataset and a test dataset. Each individual for the genetic algorithms represents a selection function of the feature vectors of the training dataset. A support vector machine is applied to each individual to evaluate the fitness value (fold classification rate) of each individual. The aim of the genetic algorithms is to search for the best individual that produces the highest fold classification rate. The best individual is then applied to the feature vectors of the test dataset and a support vector machine is built to classify protein folds based on selected features. Our experimental results on Ding and Dubchak's benchmark dataset of 27-class folds show that our approach achieves an accuracy of 71.28%, which outperforms current state-of-the-art protein fold predictors.

SubChlo: predicting protein subchloroplast locations with pseudo-amino acid composition and the evidence-theoretic K-nearest neighbor (ET-KNN) algorithm

The chloroplast is a type of plant specific subcellular organelle. It is of central importance in several biological processes like photosynthesis and amino acid biosynthesis. Thus, understanding the function of chloroplast proteins is of significant value. Since the function of chloroplast proteins correlates with their subchloroplast locations, the knowledge of their subchloroplast locations can be very helpful in understanding their role in the biological processes. In the current paper, by introducing the evidence-theoretic K-nearest neighbor (ET-KNN) algorithm, we developed a method for predicting the protein subchloroplast locations. This is the first algorithm for predicting the protein subchloroplast locations. We have implemented our algorithm as an online service, SubChlo (http://bioinfo.au.tsinghua.edu.cn/subchlo). This service may be useful to the chloroplast proteome research.